四自由度载人高逼真运动模拟平台的研究

发布时间：2018-08-01 18:07

【摘要】：运动模拟平台是一种用于模拟飞行器及其他交通工具中驾驶人员动态感受的地面模拟仿真设备,能够让模拟平台上的驾乘人员感受到真实飞行过程中相近的瞬时过载及姿态变化。但传统的六自由度并联运动模拟平台及其经典洗出算法难以实现对真实瞬时过载及姿态变化的高逼真度模拟。因此基于实现高逼真度动态模拟的目标,完成了新型运动模拟平台机械结构的设计工作,并对高逼真动态模拟算法及运动模拟平台控制系统等一系列问题展开了研究。提出了一种四自由度串联机构形式的运动模拟平台设计方案。基于设计技术协议参数,利用Solidworks,Adams,Workbench,ANSYS等软件完成了运动模拟平台的机械结构设计及优化、动力学仿真分析及电机力矩校核、有限元仿真分析及静强度校核、刚柔耦合动力学仿真分析及动强度校核,在追求模拟平台轻量化的同时,保证了关键部件的动态强度。动态模拟算法对运动模拟平台模拟逼真度拥有决定性的影响。基于人体前庭器官数学模型,建立了动态模拟算法逼真度的客观评价体系,为调整滤波器参数提供了依据。对动态模拟算法中各个环节的组成进行了具体分析,选择并优化了各环节传递函数形式及滤波器参数。考虑到固定参数的动态模拟算法对动态细节信息描述不准确等缺点,基于模糊控制理论设计了能够根据输入信号频率实时调节动态模拟算法滤波器参数的新型模糊动态模拟算法。以STM32微控制器为核心,设计了丰富的外围硬件电路。并根据控制要求,设计了控制系统软件部分,实现了电机位置控制功能,优化了电机加减速曲线,结合串口通信完成了上位机通过串口对运动模拟平台各轴电机的运动控制功能。利用卡尔曼滤波对MPU6050的原始姿态数据进行了融合数据处理,得到了动态精度较高、噪声干扰小的运动模拟平台实时运动姿态数据,为实现控制算法性能的闭环评估奠定了基础。搭建了运动模拟平台物理样机,制作了控制系统PCB板实物。以汽车驾驶为例,分别使用了经典洗出算法和新型模糊动态模拟算法在不同的路况下进行了直线加减速、弯道加减速、综合路况等测试。通过与真实汽车驾驶运动数据进行对比,结果表明提出的新型模糊动态模拟算法相比经典洗出算法对动态细节信号描述更加准确,输出运动的幅值差与相位差更小,实现了更高逼真度的动态模拟。
[Abstract]:The motion simulation platform is a ground simulation equipment used to simulate the dynamic feelings of drivers in aircraft and other vehicles. Drivers on the simulation platform can feel the similar instantaneous overload and attitude change during the real flight. However, the traditional parallel motion simulation platform with six degrees of freedom and its classical washing out algorithm are difficult to achieve high fidelity simulation of real instantaneous overload and attitude change. Therefore, based on the goal of realizing high fidelity dynamic simulation, the design of the mechanical structure of a new motion simulation platform is completed, and a series of problems such as high-fidelity dynamic simulation algorithm and motion simulation platform control system are studied. A motion simulation platform with four degrees of freedom in series is proposed. Based on the parameters of the design protocol, the mechanical structure design and optimization of motion simulation platform, dynamic simulation analysis, motor torque check, finite element simulation analysis and static strength check are completed by using SolidworksAdamsn Workbench ANSYS and other softwares. The dynamic simulation analysis and dynamic strength checking of rigid-flexible coupling ensure the dynamic strength of key components while pursuing the lightweight of the simulation platform. Dynamic simulation algorithm has a decisive effect on simulation fidelity of motion simulation platform. Based on the mathematical model of human vestibular organ, an objective evaluation system of the fidelity of the dynamic simulation algorithm is established, which provides the basis for adjusting the filter parameters. The composition of each link in the dynamic simulation algorithm is analyzed in detail, and the form of transfer function and filter parameters are selected and optimized. Considering the shortcomings of dynamic simulation algorithm with fixed parameters, such as inaccurate description of dynamic detail information, Based on the fuzzy control theory, a new fuzzy dynamic simulation algorithm is designed, which can adjust the filter parameters of the dynamic simulation algorithm in real time according to the frequency of the input signal. With STM32 microcontroller as the core, a rich peripheral hardware circuit is designed. According to the control requirements, the software part of the control system is designed, the motor position control function is realized, and the acceleration and deceleration curve of the motor is optimized. Combined with serial communication, the motion control function of each axis motor of the motion simulation platform is completed by the host computer through the serial port. Using Kalman filter to process the original attitude data of MPU6050, the real-time motion attitude data of motion simulation platform with higher dynamic precision and less noise interference are obtained, which lays a foundation for closed-loop evaluation of control algorithm performance. The physical prototype of the motion simulation platform is built and the PCB board of the control system is made. Taking automobile driving as an example, the classical washing out algorithm and the new fuzzy dynamic simulation algorithm are used to test the linear acceleration and deceleration, the curve acceleration and deceleration, and the comprehensive road condition under different road conditions. By comparing with the real driving motion data, the results show that the new fuzzy dynamic simulation algorithm is more accurate than the classical washing out algorithm in describing the dynamic detail signal, and the amplitude difference and phase difference of the output motion are smaller. A higher fidelity dynamic simulation is realized.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TH122

【参考文献】